1. Volume 12, Issue 4, Pages 743-756 (April 2019)
WDR5, BRCA1, and BARD1 Co-regulate the DNA Damage Response and Modulate the Mesenchymal-to-Epithelial Transition during Early Reprogramming 
Georgina Peñalosa-Ruiz, Vicky Bousgouni, Jan P. Gerlach, Susan Waarlo, Joris V. van de Ven, Tim E. Veenstra, José C.R. Silva, Simon J. van Heeringen, Chris Bakal, Klaas W. Mulder, Gert Jan C. Veenstra 
Stem Cell Reports 
Volume 12, Issue 4, Pages (April 2019)
DOI: /j.stemcr
Copyright © 2019 The Authors Terms and Conditions

2. Stem Cell Reports 2019 12, 743-756DOI: (10.1016/j.stemcr.2019.02.006)
Copyright © 2019 The Authors Terms and Conditions

3. Figure 1 High-Throughput Analysis of the Early Phase of Reprogramming
(A) Experimental design of high-content imaging siRNA screen.
(B) Representative image showing an immunofluorescence of reprogramming intermediates at day 6 stained for pluripotency markers CDH1 and SALL4 with DAPI counterstain. Scale bars, 50 μm.
(C) Comparison of colony phenotypes of control, siMyc and siOct4 cells at reprogramming day 6, stained for SALL4 and CDH1. Scale bars, 100 μm (images on the right, 4× zoom-in of inset).
(D) siRNAs in the whole screen ranked from low to high Z scores, based on the number of colonies. Positive controls are highlighted in colors. Each siRNA represents the average Z score from four replicates (independent transfections in the same experiment). There are 350 siRNAs because controls are included in the rank.
See also Figure S1 and Tables S1–S3.
Stem Cell Reports  , DOI: ( /j.stemcr )
Copyright © 2019 The Authors Terms and Conditions

4. Figure 2
High-Content Microscopy Screen Reveals Five Major Phenotypes of Colony Formation
(A) An average Z score for selected high-content features was calculated from quadruplicate samples (independent siRNA transfections from same experiment) and represented in a heatmap. Features are clustered by Euclidean distance and rows are clustered by K-means. Cluster number (left) and hits of the screen and the controls (right, number in brackets) are indicated.
(B) Examples of knockdowns showing different phenotypes. Scale bars, 200 μm.
(C) Pluripotency-associated hits were selected based on a combination of a probability prediction by machine learning based on known reprogramming facilitators (x), and a correlation analysis with the positive and negative controls (y). Selected top-hits are colored according to the cluster number (A; cf. Tables S3 and S4, Figure S2). Each data-point represents average of four replicates (A).
Stem Cell Reports  , DOI: ( /j.stemcr )
Copyright © 2019 The Authors Terms and Conditions

5. Figure 3
Transcriptome-Based Secondary Screening and Time Course Expression Analysis
(A) Overview of experimental design. Selected hits (30) and controls were transfected in triplicate and cultured until reprogramming day 6. The transcriptomes were analyzed together with a time series of control cells.
(B) siRNA-to-siRNA Pearson correlation heatmaps based on transcriptomes.
(C) Scatterplot representing pairwise siRNA correlations of gene expression values (x axis) and high-content imaging features (y axis). siRNA pairs with highest correlations in both approaches are highlighted.
(D) Progression of reprogramming in knockdown cells (black) compared with cells of the time course (orange), based on PCA analysis of the transcriptomes and the projection of all data points on a curve fitted to the time course.
(E) Boxplots representing log transformed and normalized gene expression values from the CELSeq2 time course dataset show expression of different groups of genes (Experimental Procedures, Table S5, Figures S3 and S4). The box represents the interquartile range and the line is the median. (B)–(E) Averages of three independent RNA sequencing replicates.
Stem Cell Reports  , DOI: ( /j.stemcr )
Copyright © 2019 The Authors Terms and Conditions

6. Figure 4
BRCA1, BARD1, and WDR5 Functionally Interact in Early Reprogramming
(A) Gene expression of Brca1, Bard1, and Wdr5 measured by qRT-PCR. Fold change was calculated relative to MEFs (day 0) gene expression. Each data point represents the mean value ± SD of a duplicate from the same experiment.
(B) Dot plot showing the number of SALL4-positive colonies measured by in-cell western in control and Brca1, Bard1, and Wdr5 knockdowns at day 6. Replicates are independent transfections from the same experiment. Statistical significance determined by one-way ANOVA (∗∗∗p < ).
(C) SALL4-colony ratios of the single and double knockdowns compared with the non-targeting (nt) control, measured by in-cell western at day 6. Functional interaction is determined by comparing the mean difference in double knockdown colony ratios: observed (Obs.) versus expected (Exp.). Replicates are independent transfections of an experiment performed at least twice. Statistical significance (∗p < 0.05) was calculated with two-tailed t-test.
(D) Dot plots to show Wdr5, Brca1, or Bard1 gene expression as counts per million (CPM) reads in siBard1, siBrca1, siWdr5, and nt control. (B)–(D) Each dot or data point represents a replicate and the lines represent mean ± SD.
Stem Cell Reports  , DOI: ( /j.stemcr )
Copyright © 2019 The Authors Terms and Conditions

7. Figure 5
WDR5, BARD1, and BRCA1 Are Functionally Connected in the DNA Damage Response Pathway
(A) Representative FACS histograms showing the cell distribution with log-intensity of γH2A.X in reprogramming populations measured in different conditions (white, nt; purple, siRNA). Each sample was measured in at least three independent experiments.
(B) Dot plot representing the quantification of γH2AX-positive cells in each condition. Data points correspond to biological replicates from independent experiments. Statistical significance was determined by one-way ANOVA (∗p < 0.05, ∗∗p < 0.005, ∗∗∗p < ).
(C) Confocal images of reprogramming cells at day 3, stained for γH2A.X (green) OCT4 (red), and DNA (DAPI, blue). Scale bars, 100 μm (left-middle). Zoom-in (right): magnification from inset, showing characteristic γH2A.X foci in all samples except nt control.
Stem Cell Reports  , DOI: ( /j.stemcr )
Copyright © 2019 The Authors Terms and Conditions

8. Figure 6
Wdr5, Brca1, and Bard1 Depletion Affects Expression Profiles of MET and DNA Repair Genes
(A) Volcano plots derived from two independent RNA sequencing replicates for siBard1 (left), siBrca1 (middle), and siWdr5 differential gene expression at reprogramming day 3. Highlight: differentially expressed genes (log2-fold change ≥1, adjusted p < 0.05).
(B) Bubble plot with enriched gene ontology terms (upregulated genes, orange; downregulated genes, blue) at day 3. Bubble sizes represent the number of genes.
(C) Gene set enrichment analysis for DNA repair by homologous recombination (HR) comparing siWdr5 versus control transcriptomes (left). NES, normalized enrichment score; FDR, false discovery rate. Heatmap for siBrca1, siBard1, and siWdr5 samples showing genes for DNA repair by HR (log2-ratio relative to control).
(D) Gene expression (RNA sequencing CPM) of signaling genes (magenta) and cell proliferation markers (yellow) in control, siBrca1, siBard1, and siWdr5 cells (each data point represents individual RNA sequencing replicates from independent experiments; bars, average ± SD).
(E) Heatmap representing the log2-ratio of mesenchymal and epithelial gene expression of the three knockdowns relative to control.
See Figure S5 and Table S6.
Stem Cell Reports  , DOI: ( /j.stemcr )
Copyright © 2019 The Authors Terms and Conditions